parisc: Strip upper 32 bit of sum in csum_ipv6_magic for 64-bit builds
PATCH v2 1/1 parisc: Strip upper 32 bit of sum in csum_ipv6_magic for 64-bit builds
IPv6 checksum tests with unaligned addresses on 64-bit builds result in unexpected failures.
Expected expected == csum_result, but
expected == 46591 (0xb5ff)
csum_result == 46381 (0xb52d)with alignment offset 1
Oddly enough, the problem disappeared after adding test code into the beginning of csum_ipv6_magic().
As it turns out, the 'sum' parameter of csum_ipv6_magic() is declared as __wsum, which is a 32-bit variable. However, it is treated as 64-bit variable in the 64-bit assembler code. Tests showed that the upper 32 bit of the register used to pass the variable are _not_ cleared when entering the function. This can result in checksum calculation errors.
Clearing the upper 32 bit of 'sum' as first operation in the assembler code fixes the problem.
---
v2: Updated subject, clarified that the problem only affects 64-bit builds,
added Helge's Acked-by: tag.
arch/parisc/include/asm/checksum.h | 1 + 1 file changed, 1 insertion(+)
arch/parisc/include/asm/checksum.h +1 -0
@@ -126,6 +126,7 @@ static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,** Try to keep 4 registers with "live" values ahead of the ALU. */ +" depdi 0, 31, 32, %0\n" /* clear upper half of incoming checksum */ " ldd,ma 8(%1), %4\n" /* get 1st saddr word */ " ldd,ma 8(%2), %5\n" /* get 1st daddr word */ " add %4, %0, %0\n"
Signature · 1 line
2.39.2
Quoted 17 lines · Guenter Roeck
IPv6 checksum tests with unaligned addresses on 64-bit builds result in unexpected failures.
Expected expected ==
csum_result, butexpected == 46591 (0xb5ff) csum_result == 46381 (0xb52d)with alignment offset 1
Oddly enough, the problem disappeared after adding test code into the beginning of
csum_ipv6_magic().As it turns out, the 'sum' parameter of
csum_ipv6_magic()is declared as__wsum, which is a 32-bit variable. However, it is treated as 64-bit variable in the 64-bit assembler code. Tests showed that the upper 32 bit of the register used to pass the variable are _not_ cleared when entering the function. This can result in checksum calculation errors.
Quoted 6 lines
Clearing the upper 32 bit of 'sum' as first operation in the assembler code fixes the problem.
Acked-by Helge Deller <[email protected]>Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")Signed-off-by Guenter Roeck <[email protected]>
applied.
Thanks!
Helge
Quoted 19 lines
---
v2: Updated subject, clarified that the problem only affects 64-bit builds,
added Helge's Acked-by: tag.arch/parisc/include/asm/checksum.h | 1 + 1 file changed, 1 insertion(+)arch/parisc/include/asm/checksum.h +1
-0@@ -126,6 +126,7 @@ static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,** Try to keep 4 registers with "live" values ahead of the ALU. */ +" depdi 0, 31, 32, %0\n" /* clear upper half of incoming checksum */ " ldd,ma 8(%1), %4\n" /* get 1st saddr word */ " ldd,ma 8(%2), %5\n" /* get 1st daddr word */ " add %4, %0, %0\n"